Notebooks

Contents

• Processing
  • Downtime categories
  • Finding the normal pitch angle threshold
  • Plot power curves for turbine 1 with data filters
  • Plot power curves for turbine 2 with data filters
  • Plot power curves for all 25 turbines
  • Process SCADA and downtime data
• Optimisation
  • Labelling data and comparing classification algorithms
  • Decision trees: finding optimal criterion (‘gini’ or ‘entropy’)
  • k-NN: finding optimal weight function (‘distance’ or ‘uniform’)
  • Comparing random forest classifier performance
  • Random forests: finding optimal criterion (‘gini’ or ‘entropy’)
• Results
  • Results for decision tree classifier
  • Results for decision tree classifier
  • Decision tree classifier results
  • Results for k-NN classifier
  • k-NN classifier results
  • Results for random forest classifier
  • Results for random forest classifier
  • Random forest classifier optimisation
  • Results for random forest classifier
  • Random forest classifier results
  • Plots of results